AISLE: an automatic volumetric segmentation method for the study of lung allometry

Area of Science:

• Medical Imaging
• Radiology
• Computational Anatomy

Background:

• Accurate lung segmentation is crucial for quantitative analysis and building statistical atlases.
• Existing semi-automatic methods require manual interaction, limiting efficiency.

Purpose of the Study:

• To develop a fully automatic segmentation method for volumetric CT datasets.
• To support the construction of a statistical atlas for studying allometric laws of the lung.

Main Methods:

• Developed AISLE (Automated ITK-Snap based on Level-set), a novel method leveraging level-set implementation.
• Utilized ITK-Snap, an existing semi-automatic segmentation program, for the core algorithm.
• Enabled segmentation of the lung field without human interaction, with options for intermediate graphical results.

Main Results:

• Preliminary experiments demonstrated accurate lung segmentation using the AISLE method.
• The volumetric overlap metric was used for comparison against radiologist-performed ground-truth segmentation.
• The method achieved high accuracy in segmenting lung fields from volumetric CT data.

Conclusions:

• The developed AISLE method provides a fully automatic and accurate solution for lung segmentation in CT datasets.
• This automated approach facilitates the creation of statistical lung atlases for allometric law research.
• AISLE shows potential for improving efficiency and consistency in lung image analysis.